# CVE-2026-115726: Aider command execution via malicious config file

Source: https://junglewise.ai/threats/cve-2026-115726-aider-through-0-86-2-contains-a-command-execution-vulnerability  
Publisher: Junglewise Threat Intelligence, https://junglewise.ai/threats  
Published: 2026-10-23. Updated: 2026-10-23  
License: CC BY 4.0 with a link back

CVE-2026-115726 · Severity: high · CVSS 7.8 · Published 2026-10-23

Aider is an AI-assisted coding tool that automatically loads configuration files. Attackers can place a malicious .aider.conf.yml file in a repository that executes arbitrary shell commands when the AI model replies or certain editor functions are used, potentially compromising the developer's system with their full privileges.

Technologies: [Aider-AI Aider](https://junglewise.ai/threats/technologies/aider). Vendors: [Aider-AI](https://junglewise.ai/threats/vendors/aider-ai).

## Quick answers

**Is CVE-2026-115726 exploited in the wild?** No. Junglewise Threat Intelligence has no report of CVE-2026-115726 being exploited in the wild.

**How severe is CVE-2026-115726?** High severity, with a CVSS score of 7.8, published 23 October 2026.

**Which products and versions are affected?** Aider-AI Aider through 0.86.2.

**Is there a fix for CVE-2026-115726?** No fixed version is recorded yet. Restrict repository access to trusted sources and avoid cloning or working with repositories from untrusted origins until upgrading. If you run it, Junglewise drafts a mitigation that fits your setup and applies it once you approve, with senior security engineers behind it.

**How is CVE-2026-115726 attacked?** It is attacked with local access or over the network, in a software library.

**Am I affected by CVE-2026-115726?** Check whether you run Aider-AI Aider at an affected version. Junglewise does that check for you around the clock: connect your code, cloud and domains, and it matches new vulnerabilities like this one against what you actually run and tells you when one reaches you.

## How to fix


Restrict repository access to trusted sources and avoid cloning or working with repositories from untrusted origins until upgrading.

## Technical details

The vulnerability exists in how Aider auto-loads .aider.conf.yml configuration files without a user trust prompt. An attacker can set the "notifications-command" or "editor" configuration values to arbitrary shell commands, which are then executed via Python's subprocess with shell=True after model responses or /editor command invocation. This requires local or network access to place the malicious config in the repository, but no authentication or user interaction beyond normal tool usage is needed to trigger execution.

## Affected products

- Aider-AI Aider through 0.86.2

## Timeline

- 2026-10-23: disclosed

## References

- [https://github.com/Aider-AI/aider/blob/v0.86.2/aider/editor.py](https://github.com/Aider-AI/aider/blob/v0.86.2/aider/editor.py)
- [https://github.com/Aider-AI/aider/blob/v0.86.2/aider/io.py](https://github.com/Aider-AI/aider/blob/v0.86.2/aider/io.py)

## Related threats

- [aider arbitrary command execution via .aider.conf.yml configuration file](https://junglewise.ai/threats/cve-2026-85674-aider-arbitrary-command-execution-via-aider-conf-yml)
- [Aider-AI Aider SSRF in URL scraping component](https://junglewise.ai/threats/cve-2026-10177-aider-ai-aider-ssrf-in-url-scraping-component)
- [Aider-AI Aider SQL injection in Code Generation Workflow](https://junglewise.ai/threats/cve-2026-10176-aider-ai-aider-sql-injection-in-code-generation-workflow)
- [Aider-AI Aider code injection in Architect Mode](https://junglewise.ai/threats/cve-2026-10175-aider-ai-aider-code-injection-in-architect-mode)
- [Aider-AI Aider protection mechanism failure in Pre-commit Hook Handler](https://junglewise.ai/threats/cve-2026-10174-aider-ai-aider-protection-mechanism-failure-in-pre-commit-hook)